A Clinical Comparative Study of 3‐Dimensional Accuracy between Digital and Conventional Implant Impression Techniques

Purpose To evaluate the accuracy of a digital implant impression technique using a TRIOS 3Shape intraoral scanner (IOS) compared to conventional implant impression techniques (pick‐up and transfer) in clinical situations. Materials and Methods Thirty‐six patients who had two implants (Implantium, internal connection) ranging in diameter between 3.8 and 4.8 mm in posterior regions participated in this study after signing a consent form. Thirty‐six reference models (RM) were fabricated by attaching two impression copings intraorally, splinted with autopolymerizing acrylic resin, verified by sectioning through the middle of the index, and rejoined again with freshly mixed autopolymerizing acrylic resin pattern (Pattern Resin) with the brush bead method. After that, the splinted assemblies were attached to implant analogs (DANSE) and impressed with type III dental stone (Gypsum Microstone) in standard plastic die lock trays. Thirty‐six working casts were fabricated for each conventional impression technique (i.e., pick‐up and transfer). Thirty‐six digital impressions were made with a TRIOS 3Shape IOS. Eight of the digitally scanned files were damaged; 28 digital scan files were retrieved to STL format. A coordinate‐measuring machine (CMM) was used to record linear displacement measurements (x, y, and z‐coordinates), interimplant distances, and angular displacements for the RMs and conventionally fabricated working casts. CATIA 3D evaluation software was used to assess the digital STL files for the same variables as the CMM measurements. CMM measurements made on the RMs and conventionally fabricated working casts were compared with 3D software measurements made on the digitally scanned files. Data were statistically analyzed using the generalized estimating equation (GEE) with an exchangeable correlation matrix and linear method, followed by the Bonferroni method for pairwise comparisons (α = 0.05). Results The results showed significant differences between the pick‐up and digital groups in all of the measured variables (p < 0.001). Concerning the transfer and digital groups, the results were statistically significant in angular displacement (p < 0.001), distance measurements (p = 0.01), and linear displacement (p = 0.03); however, between the pick‐up and transfer groups, there was no statistical significance in all of the measured variables (interimplant distance deviation, linear displacement, and angular displacement deviations). Conclusions According to the